Download PREFIXED ADJECTIVAL PARTICIPLES

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project

Document related concepts

Causative wikipedia , lookup

Spanish grammar wikipedia , lookup

Preposition and postposition wikipedia , lookup

Old Irish grammar wikipedia , lookup

Swedish grammar wikipedia , lookup

Portuguese grammar wikipedia , lookup

Japanese grammar wikipedia , lookup

Macedonian grammar wikipedia , lookup

Old English grammar wikipedia , lookup

Proto-Indo-European verbs wikipedia , lookup

Modern Hebrew grammar wikipedia , lookup

Germanic weak verb wikipedia , lookup

Germanic strong verb wikipedia , lookup

Polish grammar wikipedia , lookup

Scottish Gaelic grammar wikipedia , lookup

Udmurt grammar wikipedia , lookup

Ojibwe grammar wikipedia , lookup

Zulu grammar wikipedia , lookup

Georgian grammar wikipedia , lookup

Old Norse morphology wikipedia , lookup

English clause syntax wikipedia , lookup

Icelandic grammar wikipedia , lookup

Latin conjugation wikipedia , lookup

Sotho verbs wikipedia , lookup

Ancient Greek verbs wikipedia , lookup

Lexical semantics wikipedia , lookup

Turkish grammar wikipedia , lookup

Ancient Greek grammar wikipedia , lookup

Lithuanian grammar wikipedia , lookup

Navajo grammar wikipedia , lookup

Serbo-Croatian grammar wikipedia , lookup

Latin syntax wikipedia , lookup

Ukrainian grammar wikipedia , lookup

Yiddish grammar wikipedia , lookup

Kannada grammar wikipedia , lookup

Pipil grammar wikipedia , lookup

Esperanto grammar wikipedia , lookup

Participle wikipedia , lookup

Transcript
PREFIXED ADJECTIVAL PARTICIPLES
In this paper, I am concerned with prefixed adjectival participles in Czech. I show that in contrast
to the most widely accepted approach to Slavic prefixes, adjectival participles can be derived by
both lexical and superlexical prefixes and that lexically and superlexically prefixed participles
can denote a target state as well as a resultant state. I propose that prefixed adjectival participles
have the following syntactic structure [AP A [PartP Part [AspP Asp [­­vP v [√P √ [pP p [PP P]]]]]]], where the
adjectival head works as a stativizer and the prefix is an incorporated preposition. The incorporated preposition introduces a state variable in its meaning, which licenses the stativizer in the
derivation.
Keywords
Participle; Adjective; Prefix; Aspect, Czech.
1. Introduction
Slavic prefixes are usually divided into two types, lexical and superlexical
(Isačenko 1962, Babko-Malaya 1999, Di Sciullo – Slabakova 2005, Svenonius
2004, Richardson 2007). In the widely accepted syntactic approach to Slavic
prefixes, the syntactic position of lexical prefixes differs from the positions of
superlexical prefixes; whereas lexicals merge in the verbal domain, superlexicals
merge outside vP/VP (e. g. Ramchand 2004, Svenonius 2004, Romanova 2006,
Gehrke 2008). Given these different positions, it has been argued that in contrast to
lexicals, superlexicals cannot occur in adjectival participles; see Romanova (2006)
for Russian past passive participles and Gehrke (2008) for Russian and Czech past
active and past passive participles. This paper offers a detailed investigation of
this matter. It is organized as follows. In Section 2, I discuss properties of -lý and
7
ČLÁNKY – ARTICLES
Abstract
64 / 2016 / 1 Petr Biskup
64 / 2016 / 1
Petr Biskup
Prefixed Adjectival Participles
-ný/-tý adjectival participles.1 In Section 3.1, I am concerned with syntactic
properties of adjectival participles – focusing primarily on target states – and in
Section 3.2 I discuss lexical and superlexical prefixes. Section 3.3 provides a sample
derivation. Section 4 concludes the paper.
2. Data
ČLÁNKY – ARTICLES
2.1 -lý participles
Imperfective intransitives derive -l participles and the past tense; see (1) for unaccusatives and (2) for unergatives.
(1) a. mrzlb.kvetl frozeblossomed
‘(He) was cold’‘(It) was blossoming’
(2) a. čarovalb.pracoval
made.magicworked
‘(He) was making magic’‘(He) was working’
The contrast between the ungrammatical (3) and the grammatical (4) with the
agent-oriented úmyslně ‘intentionally’ shows that the predicates in (1) are indeed
unaccusative and the ones in (2) unergative.
(3) a. *Úmyslněmrzl.b.*Úmyslně kvetl.
intentionallyfroze intentionallyblossomed
(4) a. Úmyslněčaroval.b.
Úmyslně pracoval.
intentionallymade.magic intentionallyworked
‘He was intentionally making ‘He was intentionally working.’
magic.’
According to Levin – Rappaport Hovav (1995), resultative constructions are possible with unaccusatives, not with unergatives, but resultative predicates can be
licensed with unergatives by a fake reflexive. Given this, data in (5) and (6) support our classification.2
(5) a. z-mrzl nakostb.roz-kvetldokrásy
from-frozeonboneapart-blossomedto
beauty
‘He was chilled to the bone.’ ‘It blossomed into a beauty.’
1
I refer to adjectival participles derived by the -ný/-tý and -lý suffixes as ‘-ný/-tý participles’ and ‘-lý
participles’ and to verbal participles derived by -l and -n/-t as ‘-l participles’ and ‘-n/-t participles’.
2
Prefixation ensures here that the result state is reached. Below, we will also see an unaccusativity
test using the formation of -lý participles.
8
Petr Biskup
Prefixed Adjectival Participles
Although the unprefixed predicates in (1) and (2) form -l participles, they do not
derive -lý participles, as illustrated below.3
The following examples show that the formation of -lý participles is sensitive to
aspectual properties. When the verbs are prefixed, some of them derive -lý participles since verbal prefixes have a perfectivizing and telicizing effect; compare
(7), (8) with (9) and (10) (see also Weiss 1977 for the claim that Polish -ły participles
can be derived only from perfective verbs and for telicity see Kratzer 1994 and
Rapp 1996).
(9) a.
(10) a.
o-mrzlý b.roz-kvetlý
about-frozeapart-blossomed
‘frostbitten’‘in blossom’
*o-čarovalýb.*vy-pracovalý
about-made.magic out-worked
The contrast between (9) and (10) shows that only unaccusative perfective/telic
verbs can derive -lý participles. For this reason, the formation of -lý participles
has been used as a diagnostic of unaccusativity (see Kosta – Frasek 2004 and
Medová 2012).
The formation of -lý participles is also dependent on the intransitivity of the
predicate. According to Lamprecht et al. (1986) and Nübler (2004), transitives
do not derive -lý participles in Modern Czech, see (11), although they form - l participles, as in (12). 4
(11) a. *roz-hrabalýb.*na-budilý
apart-raked on-woke
3
Note that cases like plynulý ‘fluent’ are true adjectives (Petr et al. 1986,1).
4
Transitives with -nou-/-nu- are exceptional since they can often form both participles, e.g. vytisknout ‘print’ derives vytisknutý and vytisklý. According to Kopečný (1962), there are also some transitives
without -nou-/-nu- that derive -lý participles, like zdědilý ‘inherited’, but they are very often archaic or
dialectal.
9
ČLÁNKY – ARTICLES
(7) a. * mrzlý b.*kvetlý
froze blossomed
(8) a. * čarovalýb.*pracovalý
made.magic worked
64 / 2016 / 1 (6) U-čaroval / u-pracoval
*(se) k smrti.
at-made.magicat-worked selfto
death
‘He made magic/worked himself to death.’
ČLÁNKY – ARTICLES
64 / 2016 / 1
Petr Biskup
Prefixed Adjectival Participles
(12) a. roz-hrabalb.na-budil
apart-rakedon-woke
‘(He) spread sth.’‘(He) energized sb.’
To conclude, only perfective/telic unaccusative predicates – which are mostly derived by prefixation – form -lý participles (with the exception of -nou-/-nu- verbs,
see note 4).
2.2 -ný/-tý participles
Intransitives do not derive -n/-t participles and the verbal passive (Karlík 2004);
see (13) for unaccusatives and (14) for unergatives. Therefore intransitives also do
not derive -ný/-tý participles; as shown in (15) and (16).
(13) a. *jemrznutb.*jekveten
isfrozen isblossomed
(14) a. *ječarovánb.*jepracován
ismade.magic isworked
(15) a. *mrznutýb.*kvetený frozen blossomed
(16) a. *čarovanýb.*pracovaný
made.magic worked
The formation of -ný/-tý participles is dependent on the presence of an accusative
object. E. g. (17) shows the transitive hrabat ‘rake’ in the -n/-t passive and -ný/-tý
participles. The verbal participles (17a,c) and the adjectival participles (17b,d) can
be perfective (prefixed) as well as imperfective (unprefixed).
(17) a. To listí bylo hrabáno.b.hrabané listí
thefoliagewasrakedrakedfoliage
‘The foliage was being raked.’ ‘The foliage that is being raked.’
c. Tolistí bylo roz-hrabáno. d. roz-hrabané listí
thefoliagewasapart-rakedapart-rakedfoliage
‘The foliage was spread.’
‘The foliage that was spread.’
In the case of verbal objects that are not marked with accusative (18a), the default
agreement is present on the -n/-t participle in the passive; see (18b). The related
-ný/-tý participle is ungrammatical (18c), in contrast to the participles in (17b,d).
(18) a. (Za-)tleskalidětem.
(behind-)clappedchildrenDAT
‘They were applauding children. / They applauded children.’
10
Petr Biskup
Prefixed Adjectival Participles
(19) a. o-čarovatb.o-čarovanýstařec
about-make.magicabout-made.magicold.man
‘bewitch’ ‘a bewitched old man’
(20) a. pře-pracovat seb.pře-pracovanýlékař
over-workselfover-workeddoctor
‘overwork’
‘an overworked doctor’
When the base verb is not transitivized by the added prefix, the -ný/-tý participle
is ungrammatical, as shown below.
(21) a. Jirkado-pracoval.b.*do-pracovaný Jirka
Jirkato-worked to-workedJirka
‘Jirka stopped working.’
(22) a. Strom roz-kvetl.b.*roz-kvetenýstrom
treeapart-blossomed apart-blossomedtree
‘The tree blossomed.’
There is also an agentivity (causativity) restriction on the formation of -ný/-tý and
-n/-t participles; consider (23) (cf. also Veselovská – Karlík 2004). Although the
experiencer verb selects an accusative object and derives an -l participle, as in
(23a), it does not derive -n/-t and -ný/-tý participles, as shown in (23b,c), independently of whether or not it is prefixed.
(23) a. Jana(za-)svědilastará rána.
JanACC(behind-)itchedoldwoundNOM ‘The old wound itched Jan. / The old wound began to itch Jan.’
b.*Janbyl (za-)svěděn.
JanNOM
was (behind-)itched
c.*(za-)svěděnýJan
(behind-)itchedJanNOM
11
ČLÁNKY – ARTICLES
The ungrammatical participles in (15) and (16) are unprefixed. When a prefix is attached to the verbs, at least some of them can derive -ný/-tý participles because the
prefix transitivizes them. The participles can be based on a non-reflexive verb, as
in (19), or on a reflexive verb, as in (20).
64 / 2016 / 1 b. Bylo(za-)tleskáno
dětem.
was(behind-)clappedchildrenDAT
‘Children were being applauded. / Children were applauded.’
c.*(za-)tleskanéděti
(behind-)clappedchildrenNOM
Petr Biskup
Prefixed Adjectival Participles
ČLÁNKY – ARTICLES
64 / 2016 / 1
To sum up, in the vast majority of cases, -ný/-tý participles are derived from agentive transitive predicates with an accusative object.
3. The analysis
3.1 Syntactic properties of adjectival participles
Since -l can attach to imperfective verbs, as in (1) and (2), the ungrammaticality of
-lý participles in (7) and (8) is not based on selectional requirements of the affix -l.
The ungrammaticality also cannot be based on some requirements of the agreement marker -ý because it can attach to imperfectives, too, as shown by hrabané
‘being raked’ in (17b). Following Kratzer (2000), I assume that there is a covert affix between the participial -l and the ending -ý that represents the adjectival head
(A) and works as a stativizer since the perfective -lý participles in (9) have a stative interpretation. Since -ný/-tý participles derived from perfective verbs also
have a stative interpretation (cf. Kopečný 1962), I also assume the stativizer in
A for perfective -ný/-tý participles. The stativizer cannot be realized by -l or -n/-t
because these affixes also derive eventive participles like hrabán ‘being raked’ and
pracoval ‘was working’ (therefore, the stativizer also cannot be attached somewhere lower in the structure). Given that unprefixed -ný/-tý participles are eventive (see e. g. hrabané), the stativizer also cannot be realized by the ending -ý.
Given that prefixes almost always turn atelic eventualities into telic ones,
I propose that prefixes introduce a state variable into the derivation. Note that
the target state operator can apply only to verbs with a ‘visible’ state (Kratzer
2000, Alexiadou et al. 2003). Then, the reason why the -lý participles in (7)
are ungrammatical in contrast to (9) is that they do not have a prefix, which
would introduce a state variable licensing the stativizer. Similarly, in the case of
-ný/-tý participles derived from eventive predicates, only the prefixed participles,
like nabuzený ‘energized’ (in contrast to buzený ‘being woken’), can have a stative
interpretation.5
As discussed in Veselovská – Karlík (2004), long endings (occurring in adjectival participles) are related to stativity and short endings (occurring in verbal
participles) to eventivity in standard Czech. Hence, I propose that A is present
only in adjectival participles and that it is the locus of the stativizer. In contrast,
verbal participles project only PartP, thus, the extended verbal projection cannot
be stativized and short endings realize φ-features of Part.
Given that -l participles are derived from all types of verbs, -l realizes any Part
head, as shown in (24a), whereas -n/-t spells out Part with features [ag, p] since
-n/-t participles can only be derived from agentive transitives; see (24b). [p] stands
5
In cases like Ta knížka je lepená ‘The book is glued’, lepená can be analyzed as an adjective (Štícha
1986).
12
Petr Biskup
Prefixed Adjectival Participles
64 / 2016 / 1 for transitivity because all verbal non-vP arguments – including the direct object –
merge in pP selected by the root (see also Section 3.2.1). I assume the copy theory of
movement and the incorporation analysis of head movement, hence the agentive
feature of the incorporated v and the categorial feature of p are also visible on the
complex head Part.
(24) a. [PartP [Part l [ ]]]
b.[PartP [Part [ag, p] n/ t [ ]]]
(25) a. *prašansněžený na střechu/střeše.
powder snowsnowedonroofACC/roofLOC
b. prašanna-sněžený na střechu/střeše.
powder snowon-snowedonroofACC/roofLOC
‘powder snow on the roof ’
We saw in 2.1 that in contrast to -l participles, -lý participles can be derived only
from unaccusatives. For this reason, I assume that A selects a complement containing the unaccusative v, as shown in (26).
(26)[AP A stat [unacc] [PartP [Part l [ ]]]]
The fact that -lý participles are only formed from perfective/telic predicates will
be derived by the semantic (in)compatibility of the stativizer with the semantic
type of its sister. Specifically, the compatibility is ensured when PartP contains
a state variable, which is introduced by a prefix.
As to -ný/-tý participles, there is no unaccusative feature on A because they are
formed from transitives. The fact that stative -ný/-tý participles are formed from
perfective/telic (prefixed) verbs will be analyzed as above in -lý participles. Merging
the stativizing A with unprefixed -ný/-tý participles will result in a semantic mismatch since these participles do not contain a state variable (except a few perfective
simplex verbs). And merging the non-stativizing A with the prefixed participles will
not derive their stative meaning. The higher structure of -ný/-tý participles looks
like (27). The difference between the two types is that A in the eventive (unprefixed)
participles does not stativize the extended verbal projection; see (27b).
(27) a. [AP Astat [PartP [Part [ag, p] n/ t [ ]]]]
b.[AP A [PartP [Part [ag, p] n/ t [ ]]]]
13
ČLÁNKY – ARTICLES
This predicts that a pP embedded under the root whose preposition does not incorporate will not enable the predicate to form -n/-t participles, in contrast to a pP
whose preposition incorporates. This prediction is borne out:
ČLÁNKY – ARTICLES
64 / 2016 / 1
Petr Biskup
Prefixed Adjectival Participles
Why is the stativizer always present in A of -lý participles but is not in -ný/-tý participles? The eventive interpretation of -lý participles seems to be blocked by the
existence of -cí participles (e. g. kvetoucí kytka ‘blossoming flower’). It is not blocked
with -ný/-tý participles since -cí participles cannot modify the accusative object,
as shown by dělající chyby ‘made mistakes’.
We saw in 2.2 that -n/-t participles receive the default agreement when the
verbal object is marked with a non-structural-accusative case, whereas -ný/-tý
participles are ungrammatical with non-structural-accusative objects. The noun
over which the adjectival participle predicates is merged outside the participle
(Belletti – Rizzi 1981, McIntyre 2012, Bruening 2014). Following Bruening
(2014), I assume that in the adjectival participle, there is a null operator in place of the
internal argument, which is forced by the head A to move to its specifier. This
movement forms a predicate of individuals.
Biskup (2015) shows that movement of the specifier of P and movement of the
prepositional complement out of defective pP (whose P does not have φ-features
and does not assign case among other things) is possible but movement of the complement out of non-defective pP is ungrammatical (i.e., preposition stranding is
not grammatical in Czech). Given this and the assumption that non-structural
cases are assigned by a covert P, the null operator generated as the complement of
the non-defective P cannot move to SpecA. Therefore, the appropriate type (<e, t>)
is not derived and the participle cannot combine with a noun, as in tleskané děti
‘clapped children’, based on the verb tleskat, taking a dative object. However, in
cases like (28a) the operator can move out of pP since P is defective. As shown in
(28b), P does not assign case (z assigns genitive) and is spelled out only as a verbal
prefix.
(28) a. z-otročená populacea.z-otročili celou populaci
from-slaved
populationfrom-slaved
entirepopulation
‘an enslaved population’‘They enslaved the entire population.’
As to -n/-t participles, they do not contain AP, with the probe moving the prepositional complement. Moreover, -n/-t participles also do not contain a null operator
since the object merges directly in pP.
3.2 Prefixes
3.2.1 Lexical prefixes
Recall from Section 1 that according to the widely accepted approach to Slavic prefixes lexicals differ from superlexicals in their syntactic positions. The proposal that
lexicals merge very low in the syntactic structure seems to be correct. Every lexical
prefix can appear in some adjectival participle. As an illustration consider (29).
14
Petr Biskup
Prefixed Adjectival Participles
(30) a. o-teklánohab.na-mrzlýled
about-flowedfooton-froze ice
‘a swollen foot’ ‘ice frozen to sth.’
c. za-padlýprstend.roz-tekláčokoláda
behind-fellringapart-flowedchocolate
‘a ring fallen behind sth.’‘a melted chocolate’
It depends on the type of the base verb whether the prefixed verb derives a -lý participle or a -ný/-tý participle. When a prefix is attached to an unaccusative, then
a -lý participle is derived, as in (30). When a prefix is attached to an unergative, as
in (29a,c), a -ný/-tý participle is derived because the verb is transitivized. When
a prefix is attached to a transitive, a -ný/-tý participle is derived because the verb
remains transitive, as in (29b,d) (cf. Schoorlemmer 1997 for Russian).
This means that prefixes can add an internal argument and this argument
competes for the same syntactic position with the complement of unaccusatives
and transitives. In addition, prefixes can also add an argumental pP. From the
constructionist point of view, this suggests that prefixes are in fact prepositions,
projecting a phrase with maximally two arguments, which is merged as the complement of the root. Therefore, I analyze lexicals as incorporated prepositions
(for other arguments, see Biskup – Zybatow 2015), following e. g. Mulder (1992),
Pitz (1994), Fowler (1996), Romanova (2006), Biskup – Putnam (2012). Then the
syntactic structure of adjectival participles with lexical prefixes projecting a pP –
which can be either transitive or intransitive – looks like (31).
(31)[AP A [PartP Part [AspP Asp [vP v [ P √ [ pP p [PP P]]]]]]]
√
Thus, in -lý participles, the root merges with a pP (the projection of the prefix),
which is in complementary distribution with the projection projected by an
empty head that introduces the internal argument of the unprefixed verb (cf.
Borer 2005). Hence, the argument structure is not augmented; maximally,
15
ČLÁNKY – ARTICLES
As shown below, these prefixes can also appear in -lý participles. This is not surprising because the participial morphemes -l and -n/-t merge in the same syntactic position in adjectival participles.
64 / 2016 / 1 (29) a. o-čarovanýstařecb.na-metené smetí
about-made.magicold.manon-sweptrubbish ‘a bewitched old man’‘a rubbish swept on sth.’
c. za-hloubanýpracovník d. roz-dělaná pračka
behind-musedworkerapart-donewashing.machine
‘a worker lost in thoughts’a disassembled washing machine’
ČLÁNKY – ARTICLES
64 / 2016 / 1
Petr Biskup
Prefixed Adjectival Participles
a prepositional phrase can be added. Since -n/-t spells out Part with agentive and
transitive properties, only an -l participle and a -lý participle can be derived.
As to prefixed -ný/-tý participles derived from unergatives, the pP adds an
unselected argument(s) to the external argument, therefore we observe transitivization of the base verb. Given the prepositional nature of prefixes and the fact
that -n/-t spells out the Part head with the features [ag, p], a -ný/-tý participle is
derived. When a prefix is attached to a transitive, the pP replaces the phrase introducing the internal argument of the base verb; hence the prefixed verb remains
transitive. As with unergatives, given that -n/-t spells out the Part head with the
features [ag, p], a -ný/-tý participle is derived with base transitives.
Examples in (29) and (30) also show that lexically prefixed -lý and -ný/-tý
participles have a resultative/change of state meaning (cf. Svenonius 2004,
Žaucer 2009); see e. g. (30b), where the telic namrzlý is related to the atelic verb
mrznout ‘freeze’. Compositionally prefixed verbs like (29b) and (30c) show that the
result state has the prepositional/prefixal meaning, which supports the view that
it is the preposition that introduces the state. Thus, by transitivization, lexicals
help the unergative base verb to derive a -ný/-tý participle and by adding the state
variable they help imperfective unaccusative base verbs to derive -lý participles.
The fact that lexicals license the stative interpretation confirms that they are
merged lower than the stativizer.
Resultant states differ from target states in that they cannot be modified by still
(Nedjalkov – Jaxontov 1988, Taraldsen – Medová 2007). (29) and (30) show
that both -lý and ‑ný/-tý lexically prefixed participles can denote a target state. For
resultant states, consider (32).
(32) a. Ten román je(*stále ještě) pře-čtený.
the novelisstillover-read
‘The novel is (*still) read through to the end.’
b. Takytka je(*stále ještě) od-kvetlá.
the flowerisstillaway-blossomed
‘The flower is (*still) withered.’
3.3.2 Superlexical prefixes
All prefixes from (29) can be used as a superlexical prefix. (33) shows that these
superlexicals can derive -ný/-tý participles; see the inchoative o- in (33a), the cumulative na- in (33b), the inchoative za- in (33c) and the inchoative roz- in (33d).
(33) a. o-slepenýřidičb.na-žehlenásukně
about-dazzleddriveron-ironedskirt
‘a dazzled driver’‘a neatly ironed skirt’
16
Petr Biskup
Prefixed Adjectival Participles
These prefixes can also appear in -lý participles; consider the inchoative o- in (34a),
the cumulative na- in (34b), the attenuative za- in (34c) and the inchoative roz- in
(34d).
64 / 2016 / 1 (34) a. o-chraptělý profesorb.na-běhléruce
about-became.hoarseprofessoron-ranhands
‘a hoarse professor’‘swollen hands’
c. za-rudlá nohad.roz-měklý sníh
behind-reddened footapart-softenedsnow
‘a reddened foot’‘softened snow’
ČLÁNKY – ARTICLES
c. za-milovanýchlapecd.roz-pracovaná kniha
behind-lovedboyapart-workedbook
‘a boy in love’‘an unfinished book’
Some of the prefixes can also have other superlexical meanings, like the attenuative na- in nakloněný ‘a little leaned’ and the attenuative o- in ovlhčený ‘a little moistened’. Other prefixes with a superlexical meaning can derive adjectival participles, too, as shown by the excessive and the repetitive pře- in přesycený ‘overfed’
and převinutý ‘rewound’, the distributive po- in pobitý ‘killed one after another’,
the completive do- in dostavěný ‘finished building’, the terminative od- in odsouhlasený ‘agreed’, the attenuative pod- in podmračený ‘a little frowned’ and the perdurative pro- in propracovaný ‘worked through’.
Generally, except the saturative za-, all superlexicals can occur in (both types
of) adjectival participles. There is no special interaction between superlexicals
and the stativizer in A. This suggests that superlexicals merge below A, just as
lexicals. Recall that the target state operator existentially binds the event variable
and projects the state variable and that it can apply only to verbs with a ‘visible’
state. Since it is the prefix that brings about the stative component, superlexicals
must merge below A. Concerning telicity, adjectival participles with superlexicals
behave like lexically prefixed participles. They can also turn atelic eventualities
into telic ones, as shown e. g. in (33b), where the accomplishment nažehlená is related to the activity žehlit ‘iron’.
As with lexicals, when a superlexical prefix is attached to an unaccusative,
a -lý participle is derived, as in (34). When the prefix is attached to an unergative,
the verb is transitivized and consequently the -ný/-tý participle can be derived,
as in (33d). And when the prefix is attached to a transitive, the derived predicate
remains transitive and a -ný/-tý participle is derived, as in (33b,c).6 Thus, at least
some superlexicals can be analyzed on a par with lexicals. These derivational facts
are accounted for as in the case of lexicals in terms of the insertion of the appro6
Oslepit does not have an unprefixed counterpart.
17
ČLÁNKY – ARTICLES
64 / 2016 / 1
Petr Biskup
Prefixed Adjectival Participles
priate morpheme into Part and the complementary distribution of the pP projected by the prefix and the pP introducing the object.
The presented data argue against the generalized distinction between lexicals
and superlexicals with respect to the possibility of the formation of adjectival participles of the widely adopted syntax based approach. It seems that ungrammatical superlexically prefixed participles should rather be analyzed on a case-by-case
basis. This is supported by the fact that one and the same superlexical can behave
differently in dependency on the prefixed verb, e. g., the inchoative roz- derives
rozpracovaná ‘unfinished’ in (33d) but does not derive a participle with bolet ‘hurt’
and the inchoative u- derives uvězněný ‘imprisoned’ with věznit ‘keep in prison’
but does not derive a participle with věřit ‘believe’.
These facts can be accounted for as follows. As to rozbolet ‘begin to hurt’, the
participle rozbolený cannot be derived because v is of the experiencer type but -n/-t
needs to be inserted into Part with features [ag, p]. And rozbolelý cannot be derived
since the stative A in -lý participle selects a complement with the unaccusative v.
As to (u)věřit ‘(begin to) believe’, it takes a dative object, which is introduced by
a covert non-defective P, hence the preposition cannot be stranded, i.e., the null
operator cannot move to SpecA and the derivation crashes.
(33) and (34) show that both -lý and -ný/-tý superlexically prefixed participles
can denote a target state. We also saw some resultant states above, e. g., ovlhčený
‘a little moistened’, převinutý ‘rewound’, pobitý ‘killed one after another’, odsouhlasený ‘agreed’. As for -lý participles, consider e. g. po-mrzlý ‘frozen one after another’.
To sum up, since the stativizer is merged in a high syntactic position, we do not
observe any peculiar interaction between it and superlexicals; almost all superlexicals can occur in adjectival participles. We have seen that by means of transitivization, superlexicals help the unergative base verb to derive a -ný/-tý participle. Superlexicals can also telicize, thereby helping base verbs to derive participles
because telicity – the state variable – is a necessary condition for the formation of
-lý participles and stative -ný/-tý participles.
3.3 The derivation
As an illustration, consider the derivation of the target state participle zapadlá in
(35).
(35) knihaza-padlázagaučem
bookbehind-fellbehindsofa
‘the book that fell behind the sofa’
The order of morphemes is derived by head incorporation to the left: [[[[za]pad]l]á]
and the form of the ending -á is determined by the case and φ-features properties
18
Petr Biskup
Prefixed Adjectival Participles
AP
λx∃s∃e[x behind ιy[sofa(y)](s) & fall(e) &
cause(s)(e) & τ(e)⊆t & τ(e)⊃⊂τ(s)]
Op1
λSλx[S(x)]
Ex. cl.
λP∃s[P(s)]
A’
∃s∃e[x behind ιy[sofa(y)](s) & fall(e) &
cause(s)(e) & τ(e)⊆t & τ(e)⊃⊂τ(s)]
A’
λs∃e[x behind ιy[sofa(y)](s) & fall(e) &
cause(s)(e) & τ(e)⊆t & τ(e)⊃⊂τ(s)]
A
λRλs∃e[R(s)(e)]
PartP
λsλe[x behind ιy[sofa(y)](s) & fall(e) &
cause(s)(e) & τ(e)⊆t & τ(e)⊃⊂τ(s)]
l Part
λRλsλe[R(s)(e)]
AspP
λsλe[x behind ιy[sofa(y)](s) & fall(e) &
cause(s)(e) & τ(e)⊆t & τ(e)⊃⊂τ(s)]
Asp
λRλsλe[R(s)(e) & τ(e)⊆t & τ(e)⊃⊂τ(s)]
vP
λsλe[x behind ιy[sofa(y)](s) & fall(e) &
cause(s)(e)]
v
λRλsλe[R(s)(e)]
√P
λsλe[x behind iy[sofa(y)](s) & fall(e) &
cause(s)(e)]
pad √
λe[fall(e)]
pP
λQλsλe[x behind ιy[sofa(y)](s) &
Q(e) & cause(s)(e)]
p
λPλQλsλe[P(s) & Q(e) & cause(s)(e)]
PP
λs[x behind ιy[sofa(y)](s)]
t1
x
za
P
λyλxλs[x behind y(s)]
19
P’
λxλs[x behind ιy[sofa(y)](s)]
DP gaučem
ιy[sofa(y)]
ČLÁNKY – ARTICLES
(36)
64 / 2016 / 1 of kniha via Agree. Since the complex head is not agentive, Part must be spelled out
as -l. The unaccusative selection feature of A is satisfied because v of zapadlá is of
the unaccusative type. Now consider the derivation (36).
Za localizes the referent of the external argument behind the referent of the
internal argument. Importantly, P also introduces the state variable. It is this state
variable that will be projected by the stativizer. The state denoted by the prepositional phrase is the result state of the prefixed predicate. The meaning of za applies
to the definite expression gaučem, derived by the Strawsonian iota operator, which
results in the meaning that the referent of x is in the state of being behind the sofa.
ČLÁNKY – ARTICLES
64 / 2016 / 1
Petr Biskup
Prefixed Adjectival Participles
Then, the meaning of P’ combines with the external argument. Since the modified noun kniha is generated outside the participle, the external argument is a null
operator. In (36) it is represented by the trace of the operator because the operator
moves to SpecAP. The trace is interpreted as a variable and its referent is determined by the assignment function. After movement of the operator to SpecAP, the
assignment function is manipulated so that the index is mapped to the variable x;
therefore I put x on all relevant nodes.
The meaning of p is crucial because it derives pP of the prefixal type. Its meaning consists of three conjuncts. The first conjunct stands for the meaning of PP, i.e.,
for the result state brought about by prefixation. The second conjunct introduces
an event with properties of the root. The third conjunct stands for the telic property of prefixes, i.e., for the fact that prefixes bring about the causative relation between the state denoted by the prepositional phrase and the eventuality denoted
by the root.7
The meaning of pP applies to the root, with the result that x is in the state of
being behind the sofa, which is caused by the event of falling. Since unaccusatives
do not have an agent, the unaccusative v is semantically empty. The perfective Asp
is standardly treated as a perfective operator, which localizes the running time of
the event within the reference time (Paslawska – von Stechow 2003). Instead
of the usual existential quantification of the event variable, I use lambda binding here because it will be the target state operator that will existentially bind it.
Because of the meaning of p, I also add the state variable and assume that the time
of e abuts the time of s. Since participles do not have T (and their own speech time),
I leave the reference time t variable free and the context will decide how it will be
interpreted.
The participial morphology (Part) is meaningless and PartP combines with the
head A, which has the meaning of Kratzer’s (2000) target state operator. It existentially closes the event argument and externalizes the state variable. Since
a prefix – i.e., a prefixal pP with a state variable – is present in the structure, the
stative meaning is derived. Without a prefixal pP, a type mismatch would arise.
Since there are no elements modifying the state in (35) and there is no reason not
to existentially bind the state variable in AP already, the existential closure closes
the state variable.
In AP with the moved operator, via the assignment function g manipulated by
Predicate Abstraction (Heim – Kratzer 1998), the index 1 is mapped to the variable x.8 I analyze the moved operator as an identity function, which returns the
<e, t> meaning of the AP (the Predicate Abstraction step is not shown in (36)). Consequently, the AP can combine with the noun kniha.
7
The variable e ranges over dynamic as well as stative eventualities.
8
The same happens in the case of movement of the operator to the edge of vP, which is not shown
in (36).
20
Petr Biskup
Prefixed Adjectival Participles
4. Conclusion
Alexiadou, Artemis – Rathert, Monika – von Stechow, Armin. 2003. Introduction: the
modules of Perfect constructions. In: Alexiadou, Artemis et al., eds. Perfect Explorations.
Berlin, New York: Mouton de Gruyter, vii–xxxviii.
Babko-Malaya, Olga. 1999. Zero Morphology: A Study of Aspect, Argument Structure, and
Case. Ph.D. thesis. Rutgers University.
Belletti, Andrea – Rizzi, Luigi. 1981. The Syntax of ne: Some Theoretical Implications.
The Linguistic Review. 1, pp. 117–154.
Biskup, Petr. 2015. On (Non-)Compositionality of Prefixed Verbs. In: Ziková, Markéta et al.,
eds. Slavic Languages in the Perspective of Formal Grammar: Proceedings of FDSL 10.5, Brno
2014. Frankfurt am Main: Peter Lang, pp. 59–78.
Biskup, Petr – Putnam, Michael. 2012. One P with two Spell-Outs: the ent-/aus-alternation
in German. Linguistic Analysis. 38, pp. 69–109.
Biskup, Petr – Zybatow, Gerhild. 2015. Verbal Prefixation in Slavonic: A Minimalist
Approach. In: Müller, Gereon et al., eds. Word-Formation. An International Handbook of
the Languages of Europe. Vol. 2. Berlin: De Gruyter Mouton, pp. 1492–1515.
Borer, Hagit. 2005. Structuring Sense. Oxford: Oxford University Press.
Bruening, Benjamin. 2014. Word Formation is Syntactic: Adjectival Passives in English.
Natural Language & Linguistic Theory. 32, pp. 363–422.
Di Sciullo, Anna-Maria – Slabakova, Roumyama. 2005. Quantification and Aspect.
In: Verkuyl, Henk et al., eds. Perspectives on Aspect. Amsterdam: Springer, pp. 61–80.
Fowler, George. 1996. An articulated theory of aspect and prefixation in Slavic. In: Toman,
Jindřich, ed. Formal Approaches to Slavic Linguistics: The College Park Meeting 1994. Ann
Arbor: Michigan Slavic Publications, pp. 97–122.
Gehrke, Berit. 2008. Ps in Motion: On the Semantics and Syntax of P Elements and Motion Events.
Ph.D. thesis. University of Utrecht.
Heim, Irene – Kratzer, Angelika. 1998. Semantics in Generative Grammar. Oxford: Blackwell.
Isačenko, Alexander. 1962. Die russische Sprache der Gegenwart. Halle: Niemeyer.
Karlík, Petr. 2004. Pasivum v češtině. Slovo a slovesnost. 65, pp. 82–113.
Kopečný, František. 1962. Základy české skladby. Praha: SPN.
21
ČLÁNKY – ARTICLES
REFERENCES
64 / 2016 / 1 We have seen that both lexical and superlexical prefixes occur in adjectival participles. I have analyzed lexicals and (some) superlexicals as incorporated prepositions projecting pP in the complement position of the root. By transitivization,
both types of prefixes help unergative verbs to derive -ný/-tý participles and by
adding the state they help unaccusative verbs to derive -lý participles. The added
state variable licenses the target state operator in the structure. Lexically and superlexically prefixed -lý and -ný/-tý participles can denote a target state as well as
a resultant state.
ČLÁNKY – ARTICLES
64 / 2016 / 1
Petr Biskup
Prefixed Adjectival Participles
Kosta, Peter – Frasek, Jens. 2004. Neakuzativita (ergativita) vs. neergativita v češtině,
polštině a jiných slovanských jazycích na rozhraní morfologie a syntaxe. In: Hladká,
Zdena – Karlík, Petr, eds. Čeština – univerzália a specifika 5. Praha: NLN, pp. 172–194.
Kratzer, Angelika. 2000. Building Statives. In: Conathan, Lisa et al., eds. Proceedings of the
Berkeley Linguistics Society 26. Berkeley: Berkeley Linguistic Society, pp. 385–399.
Lamprecht, Arnošt – Šlosar, Dušan – Bauer, Jaroslav. 1986. Historická mluvnice češtiny.
Praha: SPN.
Levin, Beth – Rappaport Hovav, Malka. 1995. Unaccusativity: At the Syntax-Lexical Semantics
Interface. Cambridge, MA: MIT Press.
McIntyre, Andrew. 2013. Adjectival passives and adjectival participles in English.
In: Alexiadou, Artemis – Schäfer, Florian, eds. Non-Canonical Passives. Amsterdam:
John Benjamins, pp. 21–42.
Medová, Lucie. 2012. Anticausatives Are Derived Unergatives. In: Ziková, Markéta –
Dočekal, Mojmír, eds. Slavic Languages in Formal Grammar. Proceedings of FDSL 8.5, Brno
2010. Frankfurt am Main: Peter Lang, pp. 291–306.
Mulder, René. 1992. The aspectual nature of syntactic complementation. Ph.D. thesis.
University of Leiden.
Nedjalkov, Vladimir – Jaxontov, Sergej. 1988. The Typology of Resultative Constructions.
In: Nedjalkov, Vladimir, ed. Typology of Resultative Constructions. Amsterdam: John
Benjamins, pp. 3–62.
Nübler, Norbert. 2004. Česká participia ze syntaktického hlediska. In: Hladká, Zdena –
Karlík, Petr, eds. Čeština – univerzália a specifika. Vol. 5. Praha: NLN, pp. 384–390.
Paslawska, A. – von Stechow, Armin. 2003. Perfect readings in Russian. In: Alexiadou,
Artemis et al., eds. Perfect Explorations. Berlin: Mouton de Gruyter, pp. 307–362.
Petr, Jan et al. 1986. Mluvnice češtiny. Vol. 1 & 2. Praha: Academia.
Pitz, Anneliesse. 1994. Nominal Signs in German. Ph. D. thesis. University of Trondheim.
Ramchand, Gillian. 2004. Time and the event: The semantics of Russian prefixes. Nordlyd.
32(2), pp. 323–361.
Rapp, Irene. 1996. Zustand? Passiv? – Überlegungen zum sogenannten „Zustandspassiv“.
Zeitschrift für Sprachwissenschaft. 15, pp. 231–265.
Richardson, Kylie. 2007. Case and Aspect in Slavic. Oxford: Oxford University Press.
Romanova, Eugenia. 2006. Constructing Perfectivity in Russian. Ph.D. thesis. University of
Tromsø.
Schoorlemmer, Maaike. 1997. The role of the internal argument in the Russian aspectual
system. In: Junghanns, Uwe – Zybatow, Gerhild, eds. Formale Slavistik. Frankfurt am
Main: Vervuert Verlag, pp. 229–239.
Svenonius, Peter. 2004. Slavic prefixes inside and outside the VP. Nordlyd. 32(2), pp. 205–253.
Štícha, František. 1986. Systémový a funkční status konstrukcí s n/ t-ovými participii
v současné češtině. Slovo a slovesnost. 47, pp. 177–185.
Taraldsen, Tarald K. – Medová, Lucie. 2007. The Czech Locative Chameleon. Nordlyd. 34(2),
pp. 200–238.
Veselovská, Ludmila – Karlík, Petr. 2004. Analytic passives in Czech. Zeitschrift für
Slawistik. 49, pp. 163–235.
22
Petr Biskup
Prefixed Adjectival Participles
Petr Biskup
Institute of Slavic Studies, Leipzig University
64 / 2016 / 1 Weiss, Daniel. 1977. Syntax und Semantik polnischer Partizipialkonstruktionen im Rahmen einer
generativ-transformationellen Sprachbeschreibung. Frankfurt am Main: Peter Lang.
Žaucer, Rok. 2009. A VP-internal/Resultative Analysis of 4 “VP-External” Uses of Slavic
Verbal Prefixes. Ph.D. thesis. University of Ottawa.
Beethovenstraße 15, Leipzig
ČLÁNKY – ARTICLES
Germany
[email protected]
23